Ovarian cancer is the 5th most common cancer in women. Mounting evidence indicates that ovarian cancer is amenable to immune therapy. Our work has shown that tumor-infiltrating effector T cells predict improved outcome, while tumor-infiltrating Treg predict shorter survival in patients with ovarian cancer. However, cancer vaccines have produced modest results to date. Work from our lab and our collaborators has shown that two important barriers in the tumor microenvironment prevent the engraftment, expansion and function of antitumor effector T cells; a) vascular endothelium in tumors erects a blood-tumor barrier preventing extravasation of effector T cells into tumors (through down-regulation of cell adhesion molecules), or killing extravasating T cells in tumors (through FasL-mediated apoptosis); b) Treg cells located in the tumor microenvironment suppress the function of effector T cells. Importantly, we recently found that these two mechanisms are interconnected; hypoxia (which drives expression of VEGF) induces also accumulation of CCR10+ Treg cells in tumors, while Treg in turn express high levels of VEGF. Thus, although VEGF blockade can attenuate the blood-tumor barrier it can also produce a rebound increase in Treg accumulation in the tumor microenvironment, preserving tolerance. In this case, suppression of Treg could deprive tumors from a critical homeostatic tolerance mechanism and could produce a powerful immunomodulatory synergism at the tumor microenvironment, allowing a relatively weak antitumor immune response induced by cancer vaccine to become clinically effective. We hypothesize that combined neutralization of VEGF and Treg can produce powerful immunomodulatory interactions to greatly enhance the efficacy of vaccine therapy. Preliminary clinical experimental data lend support to our hypothesis; in a recently completed pilot study we observed a 33% objective radiographic response and 66% clinical benefit rate in ovarian cancer patients receiving a weak vaccine (immature DCs pulsed with tumor lysate supernatants) combined with VEGF blockade and metronomic cyclophosphamide. Here we propose a phase I/II clinical study that will enable us to start carefully testing the hypothesis that re-editing the tumor microenvironment through VEGF blockade combined with Treg depletion allows tumor vaccines to achieve clinical efficacy. The following Aims are proposed: Aim 1) Conduct a pilot clinical trial of autologous whole tumor antigen-pulsed dendritic cell vaccine combined rationally with Treg depletion using denileukin diftitox (Ontak) and anti-VEGF antibody (Bevacizumab). Aim 2) Assess the safety, feasibility and clinical effects of vaccine and combinatorial immunotherapy. Aim 3) Assess the immune effects of vaccine and combinatorial immunotherapy in the periphery and at the tumor microenvironment.